File suddenly slow to edit/save/load

• Aug 21, 2021 - 16:45

I created a new file this morning added some stuff and at some point it became very slow to work with. e.g operations takes anywhere from 1/2 minute to several minutes. Its only 2 pages/34 measures, so size is not the problem. Other files are fine, so I think this file got corrupted. I restarted the machine, no change.

What can you suggest to check?

$ musescore3 --version
qt5ct: using qt5ct plugin
MuseScore3 3.2.3
$ uname -r
5.4.0-81-lowlatency
$ lsb_release -a
No LSB modules are available.
Distributor ID: Ubuntu
Description: Ubuntu 20.04.2 LTS
Release: 20.04
Codename: focal

Comments

Jojo-Schmitz

• Aug 21, 2021 - 16:53

Update to 3.6.2 (whether that helps or not)
Share that score

allelopath

• Aug 21, 2021 - 17:29

I upgraded to 3.6.2 on your suggestion. It didn't fix this problem but good to have anyway.
File attached.

Attachment	Size
slow-test.mscz	401.74 KB

Jojo-Schmitz

• Aug 21, 2021 - 17:33

whoho, takes ages to load!
It is pretty big too, 400k is quite a size, esp. for a score with only 3 staves, 36 measures, 2 pages and no images

Jojo-Schmitz

• Aug 21, 2021 - 17:41

It does have very many hairpins (more than 170000!, but shows only 3 of them).
Deleting them all (right-click one, wait..., select all, wait..., Del, wait... looooong...) might help (I'm running out of time, may have a look at this tomorrow)

[DELETED] 33759070

• Aug 21, 2021 - 19:14

I unzipped the file (expanded to 65 MB). There are 2,477,058 lines in the file, mostly the following repeated over and over and over....

 <Spanner type="HairPin">
            <HairPin>
              <subtype>0</subtype>
              <linkedMain/>
              </HairPin>
            <next>
              <location>
                <measures>4</measures>
                </location>
              </next>
            </Spanner>

I removed all of these and now the score loads with a few corruptions that can probably be ignored.

Attachment	Size
faster-test.mscx	35.22 KB

Marc Sabatella

• Aug 22, 2021 - 16:35

Good job! Just a note - ignoring corruptions is never a good idea, it's pretty much guaranteed to lead to problems down the road. In this case, the corruptions may indeed be related to how these spurious hairpins appeared in the first place, and it may cause them to repeat later.

Jojo-Schmitz

• Aug 22, 2021 - 16:42

The initial score doesn’t report corruptions, only the "hand-fixed" one. But indeed those should better get fixed sooner rather than later

allelopath

• Aug 22, 2021 - 17:17

Thanks. Unfortunately, the tab notation got lost in the process. I appreciate the effort though. And also you taught me that one can go into the xml directly. Good to know.

[DELETED] 33759070

• Aug 22, 2021 - 18:40

I'm not sure how the corruptions occured, but I didn't look too carefully at the 2+ million lines I removed. I assumed they were all hairpin xml definitions.

allelopath

• Aug 22, 2021 - 17:18

Thanks for all the help. I just ended up starting over. Not sure what happened in the first place, but all is well now.